During FY2011, we extended studies of HIV Env-receptor interactions and neutralization mechanisms, with emphasis on features that distinguish the functional Env trimer from the more highly studied gp120 monomer. We also continued to develop anti-HIV agents based on the molecules involved in entry. 1) Epitope masking within the HIV-1 Env trimer. We analyzed the epitope masking mechanism(s) for two mAbs against the V3 loop. Both epitopes are inaccessible in the trimer resulting in neutralization resistance, but were rendered accessible by deletion of the V1V2 region thereby conferring neutralization sensitivity. To study the masking mechanism, we employed a previously described functional complementation system that measures activity of mixed trimers containing various modifications affecting sensitivity to these mAb (V1V2 deletion, point mutation of the epitopes). We determined that for both epitopes, the V1V2 masking occurs by an intraprotomer (cis) rather than an interprotomer (trans) mechanism, i.e. V1V2 masks the V3 epitopes on the same protomer rather than on a neighboring protomer in the Env trimer. These results clearly support one of the two alternate epitope masking mechanisms that have been proposed based on structural and modeling studies. 2) Two components of the CD4-gp120 interaction. In 2011, we expanded on our mutational evidence for a specific region of CD4 that may be involved in induction/stabilization of the coreceptor binding site on gp120. First we began studies of single molecule FRET analyses to probe the CD4-gp120 binding, with the intention of probing the effects of specific sCD4 mutations on interaction with the functional trimer. Preliminary promising results have been obtained. We also designed several new constructs of CD4 (soluble and full length), as well as novel gp120 constructs to probe the unliganded (pre-CD4) structure of gp120. 3) Strategies for depletion of HIV infected cell reservoirs persisting despite HAART. Extensive focus is being applied to devise strategies whereby HIV infection can be eradicated, or at least controlled after cessation of HAART. Our approaches until recently have focused strictly on the development of anti-HIV immunotoxins to selectively kill infected cells. While these efforts are continuing, we have begun a new collaborative effort to apply an approach that would give durable killing of infected cells, including cells that might be responsible for persistent low level viremia as well as latently infected cells that become activated after cessation of HAART. The adoptive cell therapy approach involves engineering a gene encoding a chimeric antigen receptor (CAR);when transduced by retroviral vectors into patient CD8 cells ex vivo and expanded and reintroduced into the body, these cells are expected to selectively kill infected cells expressing surface HIV Env in its native form (independent of antigen processing and MHC presentation). We are employing novel targeting moieties coupled with enhanced signaling motifs coupled with improved methods for ex vivo cell manipulation to enhance the functionality and longevity of the re-introduced CAR-expressing cells. Several CAR constructs have been produced;when transduced into human CD8 cells, functional responses are obtained in culture, including stimulation of cytokine release and selective killing of Env-expressing cells. Analogous constructs with simian signaling moieties have been designed to apply this approach in the SIV-macaque model. 4) Bifunctional HIV neutralizing protein. We previously described the extremely broad and potent HIV-1 neutralizing activities of sCD4-17b, based on the ability of the chimeric protein to simultaneously bind to two distinct highly conserved domains on gp120 (i.e. the binding sites for CD4 and coreceptor). Previous analyses related to the potential application of this agent as a topical microbicide have revealed problems with proteolytic susceptibility of the linker regions attaching the sCD4 and 17b moieties, and also the VH and VL regions of the 17b SCFv. We have generated constructs with linkers reported to be much less sensitive to proteases compared to the Gly4Ser repeats used in the original constructs. These are being expressed for testing of neutralization potency and breadth.